Abstract:Reservoir heterogeneity significantly impacts the fluid flow behavior in porous media. In subsurface communi-ties including hydrocarbon or geothermal recovery, geological storage of CO2 or H2, and hydrology, transient pressure data are often used to infer the subsurface rock properties (e.g., permeability field) due to its ready availability and quick response. The accurate estimation of such properties is critical to accurately predict fluid flow in porous media. In this work, we propose a deep learning (DL) approach inspired by the Fast Marching Method (FMM), namely the Inversion Neural Network (INN), to inversely infer heterogeneous reservoir model parameters using transient pressure data. The forward model used to generate training data for the INN is established based on a semi-analytic asymptotic solution to the diffusivity equation using the diffusive time of flight (DTOF). The FMM solves the Eikonal equation for the DTOF and provides the partial derivative of pressure drop to the natural logarithm of time (hereafter pressure derivative). The pressure derivative data are then used to predict reservoir model parameters by the INN. In the homogeneous scenario, the INN architecture is a relatively simple fully connected neural network as a proof-of-concept to validate the feasibility, and it directly correlates the permeability value with the pressure derivative. In the heterogeneous scenario, as the heterogeneous permeability field is estimated based on sparse observational data of pressure derivatives, we adopt the convolutional neural network (CNN) to flexibly deal with the image-based properties, and leverage transfer learning to efficiently train a robust INN in the heterogeneous scenario. We first validated that INN can infer the homogeneous permeability fields through numerical experiments by testing root-mean-square-error (RMSE) around 1.086 md. Inspired by that, observational data with different sparsity are used to train convolutional INN and predict heterogeneous permeability fields. As the number of observational locations (nobs) increases from 3 x 3 to 48 x 48, the testing RMSE in heterogeneous scenarios decreases from 187.510 md to 18.080 md. Besides, we found that transfer learning significantly improves the predictive accuracy at low nobs, with relative error decreased by 17.25%. Finally, noisy pressure derivative data under different nobs are used to history match the heterogeneous reservoir model, and INN can infer the permeability field with a low error of 9.6% at nobs = 7 x 7. Without an iterative procedure for parameter estimation, the INN demonstrates to perform permeability inversion with CPU time in the magnitude of 9.2 x 10-4 s on a single GPU (NVIDIA Quadro P2200). The FMM-inspired INN sets up the basis for the accurate characterization of reservoir model heterogeneity by inverting pressure derivative data with both decent predictive accuracy and computational efficiency.

Multiparameter Inversion of Reservoirs Based on Deep Learning

Simultaneous prediction of multiple physical parameters using gated recurrent neural network: Porosity, water saturation, shale content

CNN-Based Network Application for Petrophysical Parameter Inversion: Sensitivity Analysis of Input–Output Parameters and Network Architecture

Well test inversion method for fractured horizontal wells based on deep neural network

Inversion Framework of Reservoir Parameters Based on Deep Autoregressive Surrogate and Continual Learning Strategy

Deep-learning-based Surrogate Model for Forward and Inverse Problems of Wave Propagation in Porous Media Saturated with Two Fluids

High-resolution Seismic Impedance Inversion Integrating the Closed-Loop Convolutional Neural Network and Geostatistics: an Application to the Thin Interbedded Reservoir

A Carbonate Reservoir Prediction Method Based on Deep Learning and Multiparameter Joint Inversion

Mixture Density Network Applied to Reservoir Parameter Inversion in Bohai Oil Field

Multidimensional Petrophysical Seismic Inversion Based on Knowledge-Driven Semi-Supervised Deep Learning

Estimation of Heterogeneous Permeability Using Pressure Derivative Data Through an Inversion Neural Network Inspired by the Fast Marching Method

Rapid Inference of Reservoir Permeability from Inversion of Travel Time Data under a Fast Marching Method Based Deep Learning Framework

Deep Learning Based Hydraulic-mechanical Simulation of Petroleum Reservoirs

Post-stack seismic inversion through probabilistic neural networks and deep forward neural networks

Reservoir Characterization Using Multi-component Seismic Data in a Novel Hybrid Model Based on Clustering and Deep Neural Network

Deep Learning Reservoir Parameters Prediction Using Bidirectional Long Short-Term Memory Network

1-D Multi-Parameter Inversion Based on Deep Neural Network for Geophysical

Reservoir parameters prediction based on spatially transferred long short-term memory network

Multi parameter inversion of electrical array lateral logging by using the neural networks

Research on Quantitative Inversion Characterization of High-Definition Electrical Imaging Logging in Oil-Based Mud Based on BPNN and MPGA-LM Algorithm

Employing Deep Learning Neural Networks for Characterizing Dual-Porosity Reservoirs Based on Pressure Transient Tests